CROSS-REFERENCE TO RELATED APPLICATION
This application is based upon and claims the benefit of priority of the prior Japanese Patent Application No. 2017-176140, filed on Sep. 13, 2017, the entire contents of which are incorporated herein by reference.
FIELD
A certain aspect of the present invention relates to an electronic component.
BACKGROUND
It has been known to improve the connection between a bump and a pad by mounting a device chip on a substrate with use of bumps with different sizes as disclosed in, for example, Japanese Patent Application Publication No. 2015-41760 (hereinafter, referred to as Patent Document 1). It has been known to eliminate the need for highly precise position control of a mounting machine by providing large solder bumps at four corners of a device chip as disclosed in, for example, Japanese Patent Application Publication No. H11-111771 (hereinafter, referred to as Patent Document 2).
SUMMARY OF THE INVENTION
According to an aspect of the present invention, there is provided an electronic component including: a substrate; four device chips each having a rectangular planar shape, the four device chips being arranged on the substrate so that a corner, which is one corner of four corners constituting a rectangle of one device chip of the four device chips, is adjacent to the corner of each of remaining three device chips of the four device chips; a first pad located on a surface at the substrate side of each of the four device chips and closest to the corner; one or more first bumps bonding the first pad to the substrate in each of the four device chips; a second pad located on the surface at the substrate side of each of the four device chips, the second pad being one of pads other than the first pad; and one or more second bumps bonding the second pad to the substrate in each of the four device chips, a sum of bonded areas between the one or more second bumps and the second pad being less than a sum of bonded areas between the first pad and the one or more first bumps.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a circuit diagram of an electronic component in accordance with a first embodiment;
FIG. 2A and FIG. 2B are a cross-sectional view and a plan view of the electronic component of the first embodiment, respectively;
FIG. 3 is a circuit diagram of a filter in the first embodiment;
FIG. 4 is a plan view of a device chip in the first embodiment;
FIG. 5A and FIG. 5B illustrate exemplary acoustic wave resonators;
FIG. 6A and FIG. 6B are schematic cross-sectional views for describing advantages in the first embodiment;
FIG. 7 is a cross-sectional view of a sample in a simulation 1;
FIG. 8A is a plan view of the upper surface of a substrate 10 in the simulation 1, and FIG. 8B is a plan view of a wiring line 22 c;
FIG. 9A and FIG. 9B are graphs of stress versus the diameter φ2 of a bump in the simulation 1;
FIG. 10 is a plan view of a sample in a simulation 2;
FIG. 11A through FIG. 11D are graphs (No. 1) of stress versus the diameter φ2 of the bump in the simulation 2;
FIG. 12A through FIG. 12D are graphs (No. 2) of stress versus the diameter φ2 of the bump in the simulation 2; and
FIG. 13 is a plan view of an electronic component in accordance with a first variation of the first embodiment.
DETAILED DESCRIPTION
In Patent Documents 1 and 2, influence of a temperature load is not considered. When a plurality of device chips are mounted on a substrate, a large stress is applied to bumps due to a temperature load or the like. Accordingly, the reliability of the bonding between the substrate and the device chip may decrease. As the diameter of the bump is increased, stress applied to the bump decreases. However, a larger diameter of the bump increases the size of the electronic component.
Hereinafter, a description will be given of embodiments with reference to the accompanying drawings.
First Embodiment
A first embodiment is an exemplary quadplexer as an electronic component. FIG. 1 is a circuit diagram of an electronic component in accordance with the first embodiment. As illustrated in FIG. 1, the quadplexer includes filters 51 through 54. The filters 51 through 54 are connected between a common terminal Ant and signal terminals T1 through T4, respectively. For example, the filters 51 and 54 are a transmit filter and a receive filter for band B1, respectively. The filters 52 and 53 are a transmit filter and a receive filter for band B2, respectively. Band B1 and band B2 are bands of Long Term Evolution (LTE) band (Evolved Universal Terrestrial Radio Access (E-UTRA) Operating Band).
The filter 51 outputs signals in the transmit band of band B1 to the common terminal Ant among high-frequency signals input to the signal terminal T1, and suppresses other signals. The filter 54 outputs signals in the receive band of band B1 to the signal terminal T4 among high-frequency signals input to the common terminal Ant, and suppresses other signals. The filter 52 outputs signals in the transmit band of band B2 to the common terminal Ant among high-frequency signals input to the signal terminal T2, and suppresses other signals. The filter 53 outputs signals in the receive band of band B2 to the signal terminal T3 among high-frequency signals input to the common terminal Ant, and suppresses other signals.
FIG. 2A and FIG. 2B are a cross-sectional view and a plan view of the electronic component of the first embodiment, respectively. FIG. 2B is a top view of a substrate 20, and indicates device chips 11 a through 11 d by dashed lines. As illustrated in FIG. 2A, the substrate 20 includes one or more insulating layers 20 a and 20 b. The insulating layers 20 a and 20 b are, for example, ceramic layers made of High Temperature Co-fired Ceramics (HTCC) or Low Temperature Co-fired Ceramics (LTCC), or resin layers.
Pads 24 and 24 a and a ring-shaped metal layer 28 are located on the upper surface of the substrate 20. Terminals 26 are located on the lower surface of the substrate 20. The terminals 26 are, for example, the common terminal Ant, the signal terminals T1 through T4, and ground terminals connected to grounds of the filters 51 through 54. An internal wiring line 22 includes via wirings 22 a and 22 b and a wiring line 22 c. The via wirings 22 a and 22 b respectively penetrate through the insulating layers 20 a and 20 b. The wiring line 22 c is located between the insulating layers 20 a and 20 b. The internal wiring lines 22 electrically connect the pads 24 and 24 a to the terminals 26. The internal wiring lines 22, the pads 24 and 24 a, the terminals 26 and the ring-shaped metal layer 28 are formed of metal layers such as, for example, copper layers, aluminum layers, gold layers, or tungsten layers.
A device chip 11 corresponds to each of the device chips 11 a through 11 d, and includes the substrate 10, a functional element 12, and pads 14 and 14 a. The functional element 12 and the pads 14 and 14 a are located on the lower surface of the substrate 10. The functional element 12 corresponds to each of the filters 51 through 54, and faces the substrate 20 across an air gap 32. The substrate 10 is, for example, a lithium tantalate substrate or a lithium niobate substrate. The pads 14 and 14 a are formed of metal layers such as, for example, copper layers, aluminum layers, or gold layers. The device chip 11 is flip-chip mounted (face-down mounted) on the substrate 20 through bumps 16 and 16 a. The bumps 16 and 16 a are, for example, gold bumps, solder bumps, or copper bumps. The bump 16 bonds the pads 24 and 14, and the bump 16 a bonds the pads 24 a and 14 a.
A sealing portion 30 is located on the substrate 20 so as to cover the device chips 11. The sealing portion 30 is bonded to the ring-shaped metal layer 28. The sealing portion 30 is formed of, for example, a metal layer made of solder or a resin layer. A lid may be located on the sealing portion 30 and the device chips 11. The functional element 12 is sealed in the air gap 32 by the sealing portion 30.
As illustrated in FIG. 2B, the four device chips 11 a through 11 d are flip-chip mounted on the substrate 20. The planar shape of each of the device chips 11 a through 11 d is rectangular (for example, rectangular or square). The filters 51 through 54 as the functional elements 12 are located on the lower surfaces of the device chips 11 a through 11 d, respectively. One of the four corners of each of the device chips 11 a through 11 d faces one of the four corners of each of the remaining device chips. For example, two sides of the four sides of the device chip 11 a face the adjacent device chips 11 b and 11 c. Opposed sides are substantially parallel.
The pads 24 include signal pads Pt2 and ground pads Pg2. The pads 24 a include common pads Pa2. In each of the four device chips 11 a through 11 d, the pad 24 a near the center of the substrate 20 is the common pad Pa2, and the pad 24 located diagonally to the common pad Pa2 is the signal pad Pt2. Other pads 24 are the ground pads Pg2. The common pads Pa2 are electrically connected to the common terminal Ant through the internal wiring lines 22. The signal pads Pt2 are electrically connected to the signal terminals T1 through T4 through the internal wiring lines 22. The ground pads Pg2 are electrically connected to ground terminals through the internal wiring lines 22.
The common pads Pa2 of the device chips 11 a through 11 d are located around the center of the substrate 20 together. This structure makes the electrical lengths between the filters 51 through 54 and the common terminal Ant approximately equal to each other. Thus, the filters 51 through 54 are easily matched. The signal pad Pt2 is located diagonally to the common pad Pa2 in each of the device chips 11 a through 11 d. This structure inhibits the transmit signal input to the signal pad Pt2 from leaking to the common pad Pa2. In addition, the transmit signal output from the common pad Pa2 is inhibited from leaking to the signal pad Pt2. Thus, the isolation from the transmit terminal to the receive terminal improves. The transmit filter and the receive filter for the same band are preferably diagonally located to each other. For example, the filters 51 through 54 are mounted to the device chips 11 a through 11 d, respectively. This configuration improves the isolation from the transmit terminal to the receive terminal of the same band.
The device chips 11 a through 11 d and the filters 51 through 54 will be described with use of the filter 51 of the device chip 11 a as an example. FIG. 3 is a circuit diagram of the filter in the first embodiment. As illustrated in FIG. 3, series resonators S1 through S5 are connected in series between the common terminal Ant and the signal terminal T1, and parallel resonators P1 through P4 are connected in parallel between the common terminal Ant and the signal terminal T1. The series resonators S1 through S5 are serially respectively divided into S1 a and S1 b, S2 a and S2 b, S3 a and S3 b, S4 a and S4 b, and S5 a and S5 b.
FIG. 4 is a plan view of the device chip in the first embodiment. FIG. 4 transparently illustrates the lower surface of the substrate 10 as viewed from above. As illustrated in FIG. 4, acoustic wave resonators 40, wiring lines 13, and the pads 14 and 14 a are located on the lower surface of the substrate 10. The acoustic wave resonator 40 includes an Inter Digital Transducer (IDT) 41 and reflectors 42. The wiring line 13 electrically connects the acoustic wave resonators 40 or electrically connects the acoustic wave resonator 40 to the pads 14 and 14 a. The bumps 16 and 16 a are bonded to the pads 14 and 14 a, respectively. The acoustic wave resonators 40 include the series resonators S1 through S5 and the parallel resonators P1 through P4. The pads 14 include a signal pad Pt1, ground pads Pg1, and dummy pads Pd1. The pad 14 a includes a common pad Pa1. The dummy pad Pd1 is not connected to the acoustic wave resonator 40 in the device chip 11, and is bonded with a bump for securing mechanical strength.
The common pad Pa1 is electrically connected to the common pad Pa2 through the bump 16 a. The signal pad Pt1 is electrically connected to the signal pad Pt2 through the bump 16. The ground pads Pg1 and the dummy pads Pd1 are connected to the ground pads Pg2 through the bumps 16.
FIG. 5A and FIG. 5B illustrate exemplary acoustic wave resonators. The acoustic wave resonator is a surface acoustic wave resonator in FIG. 5A, and is a piezoelectric thin film resonator in FIG. 5B.
As illustrated in FIG. 5A, the IDT 41 and the reflectors 42 are formed on the substrate 10. The IDT 41 includes a pair of comb-shaped electrodes 41 a facing each other. The comb-shaped electrode 41 a includes electrode fingers 41 b and a bus bar 41 c connecting the electrode fingers 41 b. The reflectors 42 are located at both sides of the IDT 41. The IDT 41 excites a surface acoustic wave on the substrate 10. The substrate 10 is a piezoelectric substrate such as, for example, a lithium tantalate substrate or a lithium niobate substrate. The IDT 41 and the reflectors 42 are formed of, for example, an aluminum film or a copper film. The substrate 10 may be bonded to a support substrate such as a sapphire substrate, an alumina substrate, a spinel substrate, a crystal substrate, or a silicon substrate. A protective film or a temperature compensation film covering the IDT 41 and the reflectors 42 may be provided. In this case, the surface acoustic wave filter and the protective film or the temperature compensation film function as the acoustic wave resonator 40 as a whole.
As illustrated in FIG. 5B, a piezoelectric film 46 is located on the substrate 10. A lower electrode 44 and an upper electrode 48 are located so as to sandwich the piezoelectric film 46. An air gap 45 is formed between the lower electrode 44 and the substrate 10. The lower electrode 44 and the upper electrode 48 excite an acoustic wave in the thickness extension mode in the piezoelectric film 46. The lower electrode 44 and the upper electrode 48 are formed of, for example, a metal film such as a ruthenium film. The piezoelectric film 46 is, for example, an aluminum nitride film. The substrate 10 is a semiconductor substrate such as, for example, a silicon substrate or a gallium arsenide, or an insulating substrate such as a sapphire substrate, an alumina substrate, a spinel substrate, or a glass substrate. As illustrated in FIG. 5A and FIG. 5B, the acoustic wave resonator 40 includes electrodes exciting an acoustic wave. Thus, the acoustic wave resonator 40 is covered with the air gap 32 so as not to restrict the vibration of the acoustic wave.
In the first embodiment, the diameters of the bumps 16 a located at the corners facing each other of the device chips 11 a through 11 d are greater than the diameters of other bumps 16. For example, the diameters of the bumps 16 and 16 a are respectively 60 μm and 120 μm.
FIG. 6A and FIG. 6B are schematic cross-sectional views for describing advantages in the first embodiment. As illustrated in FIG. 6A, the substrate 10 is bonded on the substrate 20 through the bump 16 or 16 a. A large force 60 is applied to the bump 16 or 16 a near the center of the substrate 20 due to the difference in linear thermal expansion coefficient between the substrates 20 and 10. The stress is expressed by force/cross-section area. Thus, the bump 16 a near the center of the substrate 20 to which a large thermal stress is applied is made to be larger than the bump 16. This configuration reduces the stress applied to the bump 16 a.
As illustrated in FIG. 6B, the heat from the IDT 41, which is a heat source, flows to the substrate 20 through the bump 16 or 16 a. When the cross-section area of the bump 16 a is made to be greater than the cross-section area of the bump 16, the thermal flow from the IDT 41 to the substrate 20 is changed from the thermal flow indicated by an arrow 62 to the increased thermal flow indicated by arrows 62 a. Thus, increase in temperature of the substrate 10 is reduced. In the periphery of the substrate 20, heat can be released through the sealing portion 30, but heat release through the sealing portion 30 is difficult near the center of the substrate 20. The first embodiment increases the heat release performance near the center of the substrate 20.
On the other hand, by reducing the cross-section area of the bump 16 to which a small stress is applied, the size of the electronic component is reduced.
Simulation 1
The stress applied to the bump 16 a in the first embodiment was simulated. As a first comparative example, the sample using the bump 16 with a small diameter instead of the bump 16 a was also simulated.
FIG. 7 is a cross-sectional view of a sample in the simulation 1, FIG. 8A is a plan view of the upper surface of the substrate 10, and FIG. 8B is a plan view of the wiring line 22 c. In FIG. 8A, the device chips 11 a through 11 d are indicated by dashed lines. As illustrated in FIG. 7 through FIG. 8B, the extension direction of the longer side of the substrate 20 is defined as an X direction, the extension direction of the shorter side is defined as a Y direction, and the normal direction of the upper surface of the substrate 20 is defined as a Z direction. The wiring line 22 c is located across the entire surface between the insulating layers 20 a and 20 b. The pads 24 have the same planer shape as the device chips 11 a through 11 d.
The conditions of the simulation 1 for the first embodiment are as follows.
Substrate 10: Length Lx1=2.5 mm, length Ly1=2.0 mm
Insulating layer 20 a: HTCC with a thickness t1 of 85 μm
Insulating layer 20 b: HTCC with a thickness t3 of 10 μm
Wiring line 22 c: Tungsten with a thickness t2 of 10 μm
Pad 24, Ring-shaped metal layer 28: Tungsten with a thickness t4 of 15 μm
Bump 16: Gold with a height t5 of 12.5 μm and a diameter φ1 of 60 μm
Bump 16 a: Gold with a height t5 of 12.5 μm and a diameter φ2 of 120 μm
Device chips 11 a through 11 d: 42° rotated Y-cut X-propagation lithium tantalate substrate with a thickness t6 of 0.15 mm, a length Lx2 of 1.01 mm, and a length Ly2 of 0.77 mm, where the Y direction corresponds to the X-axis orientation of the crystal orientation.
The Young's modulus, the Poisson ratio, and the linear thermal expansion coefficient of each material used in the simulation 1 are as follows. Gold has a Young's modulus of 7.72×1010 GPa, a Poisson ratio of 0.42, and a linear thermal expansion coefficient of 1.44×10−5/° C. Lithium tantalate has a Young's modulus of 2.54×1011 GPa, a Poisson ratio of 0.21, and linear thermal expansion coefficients of 0.95×10−5/° C. (the X direction), 1.61×10−5/° C. (the Y direction), and 1.07×10−5/° C. (the Z direction). HTCC has a Young's modulus of 3.1×1011 GPa, a Poisson ratio of 0.24, and a linear thermal expansion coefficient of 0.71×10−5/° C. Tungsten has a Young's modulus of 4.0×1011 GPa, a Poisson ratio of 0.28, and a linear thermal expansion coefficient of 0.44×10−5/° C. Since the pads 14 and 14 a of the device chips 11 a through 11 d are thin, they are ignored.
Calculated were the maximum value of the stresses in the Z direction in the boundary faces between the bumps 16 a and the pads 24 and the maximum value of the stresses in the Z direction in the boundary faces between the bumps 16 a and the device chips 11 a through 11 d. They were calculated for temperatures of 150° C. and −65° C. As the first comparative example, the stress when the bump 16 with a diameter of 60 μm is used instead of the bump 16 a was also calculated.
FIG. 9A and FIG. 9B are graphs of stress versus the diameter φ2 of the bump in the simulation 1. FIG. 9A illustrates the stresses in the boundary faces between the bumps 16 a and the device chips 11 a through 11 d (represented as “DEVICE CHIP SIDE”), and FIG. 9B illustrates the stress in the boundary face between the bump 16 a and the pad 24 a (represented as “SUBSTRATE SIDE”). The sizes Lx2×Ly2 of the device chips 11 a through 11 d are the same, and thus, the stress applied to the bump 16 a is the same among the device chips 11 a through 11 d.
As illustrated in FIG. 9A and FIG. 9B, the stress in the first embodiment in which the diameter φ2 of the bump 16 a is 120 μm is less than the stress in the first comparative example in which the diameter φ2 is 60 μm.
Simulation 2
As a simulation 2, the stress applied to the bump 16 a for a case where the planar shapes of the device chips 11 a through 11 d differ from each other was calculated.
FIG. 10 is a plan view of a sample in the simulation 2. The planar shapes of the device chips 11 a through 11 d differ from each other. The pads 24 have the same planar shapes as the respective device chips 11 a through 11 d. The dimensions of the device chips 11 a through 11 d were assumed as follows.
Device chip 11 a: Lx2 a=1.07 mm, Ly2 a=0.77 mm
Device chip 11 b: Lx2 b=0.94 mm, Ly2 b=0.70 mm
Device chip 11 c: Lx2 c=0.84 mm, Ly2 c=0.77 mm
Device chip 11 d: Lx2 d=1.01 mm, Ly2 d=0.60 mm
Other simulation conditions are the same as those in the simulation 1.
FIG. 11A through FIG. 12D are graphs of stress versus the diameter φ2 of the bump in the simulation 2. FIG. 11A and FIG. 11B illustrate the stress in the bump 16 a of the device chip 11 a. FIG. 11C and FIG. 11D illustrate the stress in the bump 16 a of the device chip 11 b. FIG. 12A and FIG. 12B illustrate the stress in the bump 16 a of the device chip 11 c. FIG. 12C and FIG. 12D illustrate the stress in the bump 16 a of the device chip 11 d. FIG. 11A, FIG. 11C, FIG. 12A, and FIG. 12C illustrate the stresses in the boundary faces between the bumps 16 a and the device chips 11 a through 11 d (represented as “DEVICE CHIP SIDE”), respectively. FIG. 11B, FIG. 11D, FIG. 12B, and FIG. 12D illustrate the stresses in the boundary face between the bump 16 a and the pad 24 a (represented as “SUBSTRATE SIDE”).
As illustrated in FIG. 11A through FIG. 12D, in any of the device chips 11 a through 11 d, the stress in the first embodiment in which the diameter φ2 of the bump 16 a is 120 μm is less than the stress in the first comparative example in which the diameter φ2 is 60 μm. As described above, regardless of the sizes of the device chips 11 a through 11 d, the stress applied to the bump 16 a in the first embodiment is less than that in the first comparative example.
As described in the simulations 1 and 2, the first embodiment reduces the stress applied to the bump 16 a.
First Variation of the First Embodiment
FIG. 13 is a plan view of an electronic component in accordance with a first variation of the first embodiment, and is a top view of the substrate 20, and indicates device chips by dashed lines. As illustrated in FIG. 13, bumps 16 b having the same diameters as other bumps 16 are located on the pad 24 a. Other structures are the same as those of the first embodiment, and the description thereof is thus omitted.
As in the first variation of the first embodiment, a plurality of the bumps 16 b may be located on the pad 24 a. That is, it is sufficient if the total area of the bumps 16 a or 16 b located on the single pad 24 a is greater than the area of another bump 16.
In the first embodiment and the variation thereof, one of the four corners of each of the device chips 11 a through 11 d is adjacent to one of the four corners of each of the remaining device chips. That is, the four device chips 11 a through 11 d are arranged on the substrate 10 so that a corner that is one of the four corners constituting the rectangle of one device chip is adjacent to a corner that is one of the four corners constituting the rectangle of each of the other three device chips. The pad 14 a (a first pad) is a pad closest to the adjacent corners among the pads 14 and 14 a located on the surface at the substrate 10 side of each of the device chips 11 a through 11 d. One or more bumps 16 a or 16 b (a first bump) bond the pad 14 a to the substrate 20. The pad 14 (a second pad) is one of the pads other than the pad 14 a among the pads 14 and 14 a. The bump 16 (a second bump) bonds the pad 14 to the substrate 20.
In such a structure, the area of one bump 16 in plan view is made to be less than the sum of the areas of one or more bumps 16 a or 16 b bonded to one pad 14 a in plan view. For example, a plurality of the bumps 16 may be located on a single pad 14. For example, the sum of the bonded areas between the pad 14 and one or more bumps 16 located on the single pad 14 is made to be less than the sum of the bonded areas between the pad 14 a and one or more bumps 16 a or 16 b.
This structure reduces the stress applied to the bump 16 a or 16 b located near the center of the substrate 20 to which a large stress is applied due to a temperature load. Additionally, by making the bump 16 to which a small stress is applied small, the size of the electronic component is reduced. For example, the pad 14 can be made to be smaller than the pad 14 a. Additionally, the heat release performance is improved.
When the areas of the bumps 16 and 16 a in plan view differ from each other in the Z direction, the area in plan view may be, for example, the maximum area, the minimum area, or the average area. The areas of the bumps 16 in plan view may be approximately equal to each other or different from each other to the extent of production errors. The sum of the areas of one or more bumps 16 a or 16 b (for example, the bonded areas) in plan view in the device chips 11 a through 11 d may be substantially equal to each other or different from each other to the extent of production errors.
In one of the four device chips 11 a through 11 d, the sum of the bonded areas between the pad 14 and one or more bumps 16 is preferably equal to or less than two-thirds of, more preferably equal to or less than a half of, further preferably equal to or less than one-third of the sum of the bonded areas between the pad 14 a and one or more bumps 16 a or 16 b. The bonded area between the pad 14 and one of one or more bumps 16 is preferably less than the sum of the bonded areas of the pad 14 a and one or more bumps 16 a or 16 b, more preferably two-thirds of, further preferably a half of, yet further preferably one-third of the sum of the bonded areas between the pad 14 a and one or more bumps 16 a or 16 b.
As in the first embodiment, only one bump 16 a is bonded to each pad 14 a. This structure reduces the stress applied to the bump 16 a. For example, when the bumps 16 and 16 a are formed by plating, the areas of the bumps 16 and 16 a in plan view are easily made to differ from each other.
As in the first variation of the first embodiment, a plurality of the bumps 16 b are bonded to each pad 14 a. This structure reduces the stress applied to the bump 16 b.
The bonded area between one of the bumps 16 b and the pad 14 a is approximately equal to the bonded area between the bump 16 and the pad 14 to the extent of production errors. For example, when the bumps 16 and 16 b are formed of stud bumps, it is difficult to make the areas in plan view different. Thus, the number of fabrication steps is reduced by providing the bumps 16 b having the same area as the bump 16 to the pad 14 a.
The adjacent two sides of the adjacent two device chips (for example, the device chips 11 a and 11 b) of four device chips 11 a through 11 d are substantially parallel. This structure reduces the mounting area of the device chips 11 a through 11 d.
Each of the device chips 11 a through 11 d includes an acoustic wave element as the functional element 12 facing the substrate 20 across the air gap 32. The functional element 12 may be an acoustic wave filter connected between an input terminal (for example, the common terminal or the receive terminal) and an output terminal (for example, the transmit terminal or the common terminal).
When the bump connected to the input terminal or the output terminal deteriorates, the high-frequency characteristics deteriorate. Thus, the pad 14 a is preferably connected to the input terminal or the output terminal.
The filters 51 through 54 are connected between the common terminal Ant and the signal terminals T1 through T4, respectively. In this case, the pad 14 a is preferably connected to the common terminal Ant. This configuration reduces deterioration in high-frequency characteristics.
As illustrated in FIG. 2B, the device chips 11 a and 11 d are diagonally arranged in plan view. In plan view, the device chips 11 b and 11 c are diagonally arranged. The filters 51 and 54 are a receive filter and a transmit filter for band B1 (a first band), respectively. The filters 52 and filter 53 are a receive filter and a transmit filter for band B2 (a second band) different from band B1, respectively. This configuration allows to arrange the receive terminal of the receive filter away from the transmit terminal of the transmit filter for the same band. Thus, the isolation between the transmission and reception is made to be large. In particular, when band B1 and/or B2 is a band of a Frequency Division Duplex (FDD) method, since the receive band does not overlap with the transmit band in the band, the isolation between the transmission and the reception is important. Thus, the filters 51 and 54 are preferably the receive filter and the transmit filter for band B1, and the filters 52 and 53 are preferably the receive filter and the transmit filter for the band B2.
An acoustic wave element has been described as an example of the functional element 12, but the functional element 12 may be a passive element such as an inductor or a capacitor, an active element such as a power amplifier or a switch, or a Micro Electro Mechanical Systems (MEMS) element.
A quadplexer has been described as an example using four filters, but other multiplexers may be employed. Five or more device chips may be mounted on the substrate 20.
Although the embodiments of the present invention have been described in detail, it is to be understood that the various change, substitutions, and alterations could be made hereto without departing from the spirit and scope of the invention.